1. Field of the Invention
This invention generally relates to display drivers for electro-optic displays, and in particular relates to circuitry for driving active matrix organic light emitting diode displays.
2. Description of Related Technology
Organic light emitting diodes (OLEDs) comprise a-particularly advantageous form of electro-optic display. They are bright, colourful, fast-switching, provide a wide viewing angle and are easy and cheap to fabricate on a variety of substrates. Organic LEDs may be fabricated using either polymers or small molecules in a range of colours (or in multi-coloured displays), depending upon the materials used. Examples of polymer-based organic LEDs are described in WO 90/13148, WO 95/06400 and WO 99/48160; examples of so called small molecule based devices are described in U.S. Pat. No. 4,539,507.
A basic structure 100 of a typical organic LED is shown in FIG. 1a. A glass or plastic substrate 102 supports a transparent anode layer 104 comprising, for example, indium tin oxide (ITO) on which is deposited a hole transport layer 106, an electroluminescent layer 108, and a cathode 110. The electro luminescence layer 108 may comprise, for example, a PPV (poly(p-phenylenevinylene)) and the hole transport layer 106, which helps match the hole energy levels of the anode layer 104 and electroluminescent layer 108, may comprise, for example, PEDOT:PSS (polystyrene-sulphonate-doped polyethylene-dioxythiophene). Cathode layer 110 typically comprises a low work function metal such as calcium and may include an additional layer immediately adjacent electroluminescent layer 108, such as a layer of aluminium, for improved electron energy level matching. Contact wires 114 and 116 to the anode the cathode respectively provide a connection to a power source 118. The same basic structure may also be employed for small molecule devices.
In the example shown in FIG. 1a light 120 is emitted through transparent anode 104 and substrate 102 and such devices are referred to as “bottom emitters”. Devices which emit through the cathode may also be constructed, for example by keeping the thickness of cathode layer 110 less than around 50–100 nm so that the cathode is substantially transparent.
Organic LEDs may be deposited on a substrate in a matrix of pixels to form a single or multi-colour pixellated display. A multicoloured display may be constructed using groups of red, green, and blue emitting pixels. In such displays the individual elements are generally addressed by activating row (or column) lines to select the pixels, and rows (or columns) of pixels are written to, to create a display. It will be appreciated that with such an arrangement it is desirable to have a memory element associated with each pixel so that the data written to a pixel is retained whilst other pixels are addressed. Generally this is achieved by a storage capacitor which stores a voltage set on a gate of a driver transistor. Such devices are referred to as active matrix displays and examples of polymer and small-molecule active matrix display drivers can be found in WO 99/42983 and EP 0,717,446A respectively.
FIG. 1b shows such a typical OLED driver circuit 150. A circuit 150 is provided for each pixel of the display and ground 152, Vss 154, row select 164 and column data 166 busbars are provided interconnecting the pixels. Thus each pixel has a power and ground connection and each row of pixels has a common row select line 164 and each column of pixels has a common data line 166.
Each pixel has an organic LED 156 connected in series with a driver transistor 158 between ground and power lines 152 and 154. A gate connection 159 of driver transistor 158 is coupled to a storage capacitor 160 and a control transistor 162 couples gate 159 to column data line 166 under control of row select line 164. Transistor 162 is a field effect transistor (FET) switch which connects column data line 166 to gate 159 and capacitor 160 when row select line 164 is activated. Thus when switch 162 is on a voltage on column data line 166 can be stored on a capacitor 160. This voltage is retained on the capacitor for at least the frame refresh period because of the relatively high impedances of the gate connection to driver transistor 158 and of switch transistor 162 in its “off” state.
Driver transistor 158 is typically an FET transistor and passes a (drain-source) current which is dependent upon the transistor's gate voltage less a threshold voltage. Thus the voltage at gate node 159 controls the current through OLED 156 and hence the brightness of the OLED.
The standard voltage-controlled circuit of FIG. 1b suffers from a number of drawbacks. The main problems arise because the brightness of OLED 156 is dependent upon the characteristics of the OLED and of the transistor 158 which is driving it. In general, these vary across the area of a display and with time, temperature, and age. This makes it difficult to predict in practice how bright a pixel will appear when driven by a given voltage on column data line 166. In a colour display the accuracy of colour representations may also be affected.
Two circuits which partially address these problems are shown in FIGS. 2a and 2b. FIG. 2a shows a current-controlled pixel driver circuit 200 in which the current through an OLED 216 is set by setting a drain source current for OLED driver transistor 212 using a reference current sink 224 and memorising the driver transistor gate voltage required for this drain-source current. Thus the brightness of OLED 216 is determined by the current, Icol′, flowing into adjustable reference current sink 224, which is set as desired for the pixel being addressed. It will be appreciated that one current sink 224 is provided for each column data line 210 rather than for each pixel.
In more detail, power 202, 204, column data 210, and row select 206 lines are provided as described with reference to the voltage-controlled pixel driver of FIG. 1b. In addition an inverted row select line 208 is also provided, the inverted row select line being high when row select line 206 is low and vice versa. A driver transistor 212 has a storage capacitor 218 coupled to its gate connection to store a gate voltage for driving the transistor to pass a desired drain-source current. Drive transistor 212 and OLED 216 are connected in series between a power 202 and ground 204 lines and, in addition, a further switching transistor 214 is connected between drive transistor 212 and OLED 216, transistor 214 having a gate connection coupled to inverted row select line 208. Two further switching transistors 220,222 are controlled by non-inverted row select line 206.
In the embodiment of the current-controlled pixel driver circuit 200 illustrated in FIG. 2a all the transistors are PMOS, which is preferable because of their greater stability and better resistance to hot electron effects. However NMOS transistors could also be used. This is also true of circuits according to the invention which are described below.
In the circuit of FIG. 2a the source connections of the transistors are towards GND and for present generation OLED devices Vss is typically around −6 volts. When the row is active the row select line 206 is thus driven at −20 volts and inverted row select line 208 is driven at 0 volts.
When row select is active transistors 220 and 222 are turned on and transistor 214 is turned off. Once the circuit has reached a steady state reference current Icol′ into current sink 224 flows through transistor 222 and transistor 212 (the gate of 212 presenting a high impedance). Thus the drain-source current of transistor 212 is substantially equal to the reference current set by current sink 224 and the gate voltage required for this drain-source current is stored on capacitor 218. Then, when row select becomes inactive, transistors 220 and 222 are turned off and transistor 214 is turned on so that this same current now flows through transistor 212, transistor 214, and OLED 216. Thus the current through OLED is controlled to be substantially the same as that set by reference current sink 224.
Before this steady state is reached the voltage on capacitor 218 will generally be different from the required voltage and thus transistor 212 will not pass a drain source current equal to the current, Icol′, set by reference sink 224. When such a mismatch exists a current equal to the difference between the reference current and the drain-source current of transistor 212 flows onto or off capacitor 218 through transistor 220 to thereby change the gate voltage of transistor 212. The gate voltage changes until the drain-source current of transistor 212 equals the reference current set by sink 224, when the mismatch is eliminated and no current flows through transistor 220.
The circuit of FIG. 2a solves some of the problems associated with the voltage-controlled circuit of FIG. 1b as the current through OLED 216 can be set irrespective of variations in the characteristics of pixel driver transistor 212. However the circuit of FIG. 2a is still prone to variations in the characteristic of OLED 216 between pixels, between active matrix display devices, and over time. A particular problem with OLEDs is a tendency for their light output to decrease over time, dependent upon the current with which they are driven (this may be related to the passage of electrons through the OLED). Such degradation is particularly apparent in a pixellated display where the relative brightness of nearby pixels can easily be compared. A further problem with the circuit of FIG. 2a arises because each of transistors 212, 214 and 222 must be sufficiently physically large to handle the current through OLED 216, which is equal to the Icol reference current. Large transistors are generally undesirable and, depending upon the active matrix device structure, may also obscure or prevent the use of part of a pixel's area
In an attempt to address these additional problems there have been a number of attempts to employ optical feedback to control the OLED current. These attempts are described in WO 01/20591, EP 0,923,067A, EP 1,096,466A, and JP 5-035,207 and all employ basically the same technique. FIG. 2b, which is taken from WO 01/20591, illustrates the technique, which is to connect a photodiode across the storage capacitor.
FIG. 2b shows a voltage-controlled pixel driver circuit 250 with optical feedback 252. The main components of the driver circuit 250 of FIG. 2b correspond to those of circuit 150 of FIG. 1b, that is, an OLED 254 in series with a driver transistor 256 having a storage capacitor 258 coupled to its gate connection. A switch transistor 260 is controlled by a row conductor 262 and, when switched on, allows a voltage on capacitor 258 to be set by applying a voltage signal to column conductor 264. Additionally, however, a photodiode 266 is connected across storage capacitor 258 so that it is reverse biased. Thus photo diode 266 is essentially non conducting in the dark and exhibits a small reverse conductance depending upon the degree of illumination. The physical structure of the pixel is arranged so that OLED 254 illuminates photodiode 266, thus providing an optical feedback path 252.
The photocurrent through photodiode 266 is approximately linearly proportional to the instantaneous light output level from OLED 254. Thus the charge stored on capacitor 258, and hence the voltage across the capacitor and the brightness of OLED 254, decays approximately exponentially over time. The integrated light output from OLED 254, that is the total number of photons emitted and hence the perceived brightness of the OLED pixel, is thus approximately determined by the initial voltage stored on capacitor 258.
The circuit of FIG. 2b solves the aforementioned problems associated with the linearity and variability of the driver transistor 256 and OLED 254 but exhibits some significant drawbacks in its practical implementation. The main drawback is that every pixel of the display needs refreshing every frame as storage capacitor 258 is discharged over no more than this period. Related to this, the circuit of FIG. 2b has a limited ability to compensate for ageing effects, again because the light pulse emitted from OLED 254 cannot extend beyond the frame period. Similarly, because the OLED is pulsed on and off it must be operated at an increased voltage for a given light output, which tends to reduce the circuit efficiency. Finally, capacitor 258 often exhibits non-linearities so that the stored charge is not necessarily linearly proportional to the voltage applied on column conductor 264. This results in non-linearities in the voltage-brightness relationship for the pixel as photodiode 266 passes a photocurrent (and hence charge) which is dependent upon the level of illumination it receives.